As the technology of Internet of Things (IoT) is applied in an expanding range, more and more “ubiquitous networks/IoTs” with a large span of space have emerged. To cover these ubiquitous networks with a large span of space, multiple gateways are generally required to be used simultaneously for data collection and transmission. These gateways can be interconnected using an operator network or other private networks. The complex environment of the ubiquitous networks/IoTs and usually unattended terminal devices result in unstable communications. To improve the reliability of information transmission and to satisfy the requirements of IoT applications, it is generally necessary to select cooperation nodes to achieve redundant backup and recovery of data, thereby promoting efficient network data transmission, improving network security and communication reliability, and satisfying core content of specific application requirements.
For example, the rapid economic development and the rapid expansion of cities have made the energy shortage increasingly serious. As a major reform in the energy technology, smart grids have become a current important research hotspot of countries. The smart grids ensure the safety, reliability and economy of electricity use by monitoring power loads and developing power supply plans and price adjustment schemes. Therefore, the safe operation of a grid system is required in the smart grids. When some power devices in a grid trip due to excessive loads or are damaged for natural or human factors, a smart grid needs to judge failure causes in time. However, in the existing smart grid adopting a centralized structure, the collected data is regularly reported to a management center. When a certain power device fails, it takes a period of time to discover the failure. Moreover, a failure cause cannot be found out if the collected data is lost. Therefore, the cooperation nodes need to perform data backup mutually; when one cooperation node is found to be disconnected, another cooperation node takes the place of the one cooperation node to actively report the failure and the collected data in the last backup to the management center.
In the traditional manner of selecting the cooperation nodes, the hop count or physical distance between the cooperation nodes is a main factor for consideration. However, the case where a same failure exists in multiple cooperation nodes has not been considered. Hence, two cooperation nodes are both isolated from a primary network and thus communications cannot be guaranteed and data backup and recovery cannot be conducted.